Forget being blinded by science: the Acoustics unit of the National Physical Laboratory (NPL) is more likely to deafen you with silence. Dave Robinson enters some very strange spaces.
Consider the limit of audibility, zero decibels (sound pressure level), or 0dB SPL. That’s the bottom of the scale, right? Well, no. SPL is measured relative to the threshold of hearing, so you can actually achieve negative values (it’s just your ears can’t detect them). Which begs the question, what is absolute silence? That would occur when particles cease to vibrate. How can you do that? By freezing them down to absolute zero. But you couldn’t measure that, of course, because your microphone would cease to function. Which leads to a further suggestion, that outer space is not quite a vacuum, so there’s the tiniest amount of background audio out there.
PSNE is standing in the ideal place to contemplate these questions of science and wonder: the National Physical Laboratory in Teddington. In fact, we’re standing in the hemi-anechoic test chamber, which has an ambient noise level of zero dB SPL or lower.
“People say they come in here and feel their ears collapse,” says Ian Butterworth (pictured), higher research scientist.
NPL, first open in 1900, is one of the leading centres for science and research in the UK, as well as being the National Measurement Institute. This is where the bouncing bomb was tested, where the first atomic clock was built, where radar was invented and – yes! – where Concorde was weighed in the ’70s before it was allowed to fly.
It also has a very active acoustics department, split into three groups – Sound and Air, Ultrasonics, and Underwater Acoustics – and many different projects and tasks running concurrently. One scientist is investigating the effect of ultrasound on tissue, for use in non-invasive surgery. “So if you are trying to target the liver, you don’t hit the kidney,” explains Butterworth. With a two-storey water tank and a pressurised vessel onsite, NPL can simulate the acoustics of the ocean. And elsewhere in the facility, audiometric equipment for hospitals and audiologists is being accurately calibrated. Ian Butterworth is researching acousto-optic visualization but more on that later. Quirky stuff first.
The hemi-anechoic is a large concrete box, around 6m along each edge, packed with absorbent wedges and lacking in ventilation. The box floats on springs and is surrounded by a 50cm air-conditioned ‘void’. Calibrating this void keeps the chamber at a constant temperature so the attenuation of the air does not change.
“We use the hemi for sound power measurement – say, a commercial DVD player, or a pump, or something that needs to be ‘quiet’. We place mics around the subject in a sphere, and then by integrating between measurements at each point we can calculate the total radiating power of the subject.”
Other projects undertaken here have included the analysis of the performance of environmental windshields. To do that, you need wind. But if you create wind with a fan, the fan motor creates additional noise. “So we had a record player, which we stripped down, and constructed an arrangement using that Ð spinning the windshield, instead. The windshield moves relative to the air, creating the same effect.” Testing at 78rpm, anyone? “Then we upgraded to a potter’s wheel – very quick – maybe 120rpm.”
But the ‘hemi’ is not ideal for testing microphones, or the phase calibration of speakers. “We have a fully anechoic room too.”
Butterworth heads off down the corridor.
NPL is roughly funded 50:50 from government and the private sector. Recent spending cuts have meant that government backing has been maintained but frozen – so effectively, this means a long-term cut as inflation takes its toll. Consequently, NPL is reaching out to industry more than ever.
“Someone comes and says, we have a siren that needs to be analysed. We’d do that, measure the output accurately and produce a report. The hemi-anechoic gets a lot of use for that kind of thing: companies might have the know-how, they just might not have the space.”
Through another door and we’re in the control room for the fully anechoic chamber. It’s like the hemi, but now the floor is just a 2m metal grill that retracts into the wall. A failsafe device means you can’t close the door and retract the platform together – so no one falls into the sea of absorbent wedges accidentally. Delicate, precision measurements are made here, using computer-controlled microphone positioning equipment. Tests are usually performed in the dark too – in previous experiments, low energy bulbs were found to emit a high frequency ring that affected the results. If the hemi was eerie, then being stuck in this silent space, in the dark, would soon turn you insane.
Butterworth talks of one of his current projects which uses the chamber: measuring the acousto-optic effect. In short, it involves measuring and plotting the output of a speaker using phase differences in light from a laser-vibrometer. It’s a rapid and cutting edge method whereby the performance of speakers, particularly at the crossover point, can be mapped without using physical microphones.
Off down the corridor again. Burtterworth talks of other projects in his group. The use of MEMS (MicroElectrical-Mechanical System) microphones is high on the agenda here. These are tiny, sub-millimeter devices – diaphragms on microchips – the sort of thing you find in mobile phones. An NPL initiative called DREAMSys (Distributed Remote Environmental Array & Monitoring System) involves producing noise maps with MEMS mics: many hundreds of them can be distributed over a wide area, and the data collected to produce an audio ‘signature’ of that area. The system has been used to produce maps for critical regions around Heathrow and City Airport. Butterworth says, if inexpensive test grade MEMS mics (at less than a pound a pop) can be used rather than larger, more expensive (say) B&K testing mics, then, one, the project would be a lot cheaper and, two, the mics would be less likely to be stolen while out in the field.
We’ve arrived at the third door: the reverb room. It’s the very antithesis of an anechoic chamber: there are no parallel walls, and additional reflectors have been mounted on the ceiling in order to diffuse any sound as much as possible. The decay time to silence is over 20 seconds; for low frequencies this drop to below 10 seconds due to air absorption (but that’s still impressive).The room is used for absorption measuring (fire off your test signal, plot your decay time with and without sample, calculate the absorption factor) as well as mic calibration; oh yes, and for recording choirs at Christmas or for other novelty events, because it’s easy to harmonise with your own reflections.
Butterworth leaves briefly and returns with a red balloon. “Listen to this,” he says. Standing in the middle of the reverb room, he pops the balloon. It’s like a mortar exploding. Thirty or so seconds later, it’s finally silent.
“Follow me!” he says. Back into the corridor.
Who decides the study schedule at NPL?
“A proportion of the money leftover from the main projects gets put back into exploratory research, and every six months they have a call for ideas. If you have something whacky that might not work, you get a small fund for maybe 30 days work; if you’re successful it might go on to a bigger project. There are 30 or 40 applications each time and maybe five go through. There’s a fair amount of stuff that never makes it. But you never know where something will lead unless you try.” The very essence of experimentation, in fact.
We’re back inside the hemi-anechoic chamber. Now Butterworth has a white balloon.
“Compare and contrast!” he says, popping the balloon.
It sounds like someone breaking wind discreetly. Nothing like a balloon. We both laugh.
NPL does some incredible, critical and exciting research, and thrives on imagination leading to innovation. But it’s reassuring to know the scientists there have a sense of humour too.